Influence on fixed effects of HLMs

Description

These functions calculate measures of the change in the fixed effects estimates based on the deletetion of an observation, or group of observations, for a hierarchical linear model fit using lmer.

Usage

## Default S3 method:
mdffits(object, ...)

## S3 method for class 'mer'
cooks.distance(model, group = NULL, delete = NULL, ...)

## S3 method for class 'lmerMod'
cooks.distance(model, group = NULL, delete = NULL, ...)

## S3 method for class 'lme'
cooks.distance(model, group = NULL, delete = NULL, ...)

## S3 method for class 'mer'
mdffits(object, group = NULL, delete = NULL, ...)

## S3 method for class 'lmerMod'
mdffits(object, group = NULL, delete = NULL, ...)

## S3 method for class 'lme'
mdffits(object, group = NULL, delete = NULL, ...)

Arguments

Details

Both Cook's distance and MDFFITS measure the change in the fixed effects estimates based on the deletion of a subset of observations. The key difference between the two diagnostics is that Cook's distance uses the covariance matrix for the fixed effects from the original model while MDFFITS uses the covariance matrix from the deleted model.

Value

Both functions return a numeric vector (or single value if delete has been specified) with attribute beta_cdd giving the difference between the full and deleted parameter estimates.

Note

Because MDFFITS requires the calculation of the covariance matrix for the fixed effects for every model, it will be slower.

Author(s)

Adam Loy loyad01@gmail.com

References

Christensen, R., Pearson, L., & Johnson, W. (1992) Case-deletion diagnostics for mixed models. Technometrics, 34, 38–45.

Schabenberger, O. (2004) Mixed Model Influence Diagnostics, in Proceedings of the Twenty-Ninth SAS Users Group International Conference, SAS Users Group International.

See Also

leverage.mer, covratio.mer, covtrace.mer, rvc.mer

Examples

library(lme4)
data(sleepstudy, package = 'lme4')
ss <- lmer(Reaction ~ Days + (Days | Subject), sleepstudy)

# Cook's distance for individual observations
ss.cd.lev1 <- cooks.distance(ss)

# Cook's distance for each Subject
ss.cd.subject <- cooks.distance(ss, group = "Subject")

## Not run: 
data(Exam, package = 'mlmRev')
fm <- lmer(normexam ~ standLRT * schavg + (standLRT | school), Exam)

# Cook's distance for individual observations
cd.lev1 <- cooks.distance(fm)

# Cook's distance for each school
cd.school <- cooks.distance(fm, group = "school")

# Cook's distance when school 1 is deleted
cd.school1 <- cooks.distance(fm, group = "school", delete = 1)


## End(Not run)
# MDFFITS  for individual observations
ss.m1 <- mdffits(ss)

# MDFFITS for each Subject
ss.m.subject <- mdffits(ss, group = "Subject")

## Not run: 

# MDFFITS  for individual observations
m1 <- mdffits(fm)

# MDFFITS for each school
m.school <- mdffits(fm, group = "school")

## End(Not run)

Results

R version 3.3.1 (2016-06-21) -- "Bug in Your Hair"
Copyright (C) 2016 The R Foundation for Statistical Computing
Platform: x86_64-pc-linux-gnu (64-bit)

R is free software and comes with ABSOLUTELY NO WARRANTY.
You are welcome to redistribute it under certain conditions.
Type 'license()' or 'licence()' for distribution details.

R is a collaborative project with many contributors.
Type 'contributors()' for more information and
'citation()' on how to cite R or R packages in publications.

Type 'demo()' for some demos, 'help()' for on-line help, or
'help.start()' for an HTML browser interface to help.
Type 'q()' to quit R.

> library(HLMdiag)

Attaching package: 'HLMdiag'

The following object is masked from 'package:stats':

    covratio

> png(filename="/home/ddbj/snapshot/RGM3/R_CC/result/HLMdiag/cooks.distance.Rd_%03d_medium.png", width=480, height=480)
> ### Name: mdffits.default
> ### Title: Influence on fixed effects of HLMs
> ### Aliases: cooks.distance cooks.distance.lme cooks.distance.lmerMod
> ###   cooks.distance.mer mdffits mdffits.default mdffits.lme
> ###   mdffits.lmerMod mdffits.mer
> ### Keywords: models regression
> 
> ### ** Examples
> 
> library(lme4)
Loading required package: Matrix
> data(sleepstudy, package = 'lme4')
> ss <- lmer(Reaction ~ Days + (Days | Subject), sleepstudy)
> 
> # Cook's distance for individual observations
> ss.cd.lev1 <- cooks.distance(ss)
> 
> # Cook's distance for each Subject
> ss.cd.subject <- cooks.distance(ss, group = "Subject")
> 
> ## Not run: 
> ##D data(Exam, package = 'mlmRev')
> ##D fm <- lmer(normexam ~ standLRT * schavg + (standLRT | school), Exam)
> ##D 
> ##D # Cook's distance for individual observations
> ##D cd.lev1 <- cooks.distance(fm)
> ##D 
> ##D # Cook's distance for each school
> ##D cd.school <- cooks.distance(fm, group = "school")
> ##D 
> ##D # Cook's distance when school 1 is deleted
> ##D cd.school1 <- cooks.distance(fm, group = "school", delete = 1)
> ##D 
> ## End(Not run)
> # MDFFITS  for individual observations
> ss.m1 <- mdffits(ss)
> 
> # MDFFITS for each Subject
> ss.m.subject <- mdffits(ss, group = "Subject")
> 
> ## Not run: 
> ##D 
> ##D # MDFFITS  for individual observations
> ##D m1 <- mdffits(fm)
> ##D 
> ##D # MDFFITS for each school
> ##D m.school <- mdffits(fm, group = "school")
> ## End(Not run)
> 
> 
> 
> 
> 
> dev.off()
null device 
          1 
>